Skip to content

Expanded Applications to Validate the Real-World Evidence Framework

Expanded Applications to Validate the Real-World Evidence Framework

In our previous blog posts we discussed Pilot 1.0 and Pilot 2.0 of the Real-World Evidence (RWE) Project. The goal of the RWE pilots is to create better characterization for the role real-world endpoints could play in evaluating drug effectiveness. In this blog, we will review two collaborations that formed between Friends of Cancer Research (Friends) and, IQVIA and Owkin/CHU Nantes, respectively, to build on and expand the RWE 1.0 framework.

The outcomes of Pilot 1.0 demonstrated the feasibility of and established a methodology for harmonizing definitions and methodologies across real-world datasets. The results of the pilot also substantiated the ability to identify comparable populations in different datasets using harmonized methods. Questions remain concerning validation of associated patient outcomes, such as overall survival (OS), the gold standard measure for demonstrating clinical benefit in oncology studies. However, healthcare data, particularly mortality data, is fragmented in the US healthcare system, often requiring data to be aggregated from multiple and disparate sources to generate a dataset that is fit for a specific purpose.

Through a collaboration with IQVIA and Health Data Insight CIC using data from the Cancer Analysis System (CAS) database in the United Kingdom, we sought to apply the framework established in Pilot 1.0 to confirm our previously observed associations between OS and potential proxy endpoints in a nationally sourced dataset. In France, a collaboration with Owkin/CHU Nantes tests the RWE framework in patients with advanced melanoma using a fit-for-purpose real-world database designed to promote clinical and translational research. If the findings from Pilot 1.0 can be reproduced using different data and patient populations, we can gauge the global applicability and generalizability of the RWE framework.

The CAS Pilot followed the Friends Pilot 1.0 protocol and compared the original findings to a UK Cancer Registry (CAS database) to validate and compare real-world data (RWD) in advanced non-small cell lung cancer patients (aNSCLC). CAS is a cancer registry that includes more than 99% of all cancer patients in England and contains data on patient and tumor characteristics, treatments, hospitalizations, and mortality. Through linkage with the UK’s Office for National Statistics, CAS provides vital status information on all patients, including death records and death date, which was critical in validating the derived endpoint from the original pilot project. Results of the CAS study also demonstrated a high level of correlation between real-world OS and other real-world endpoints, indicating the potential use of RWD as a proxy endpoint for OS in real-world studies.

Similarly, the Owkin/CHU Nantes Pilot investigated the application of the Friends’ RWE framework in patients with melanoma, which provided an opportunity to align on data quality, standards, and inform expansion of the framework into other disease settings. Ultimately, the applicability of the framework in aNSCLC and melanoma, supports further development of the framework as the structure for future studies.

By expanding the RWE 1.0 framework to assess more diverse populations and healthcare settings, these studies provide more widespread support for the inclusion of RWD into regulatory and payment structure decisions. The consistent results observed between these datasets and disease settings further illustrate the integrity of the framework established through the RWE pilots. Our next blog post will focus on extended results from RWE Pilot 2.0, in which we compare the RWD and results to outcomes for front-line treatment regimens in clinical trials.


Conference FDA News Research